- Add UUID fields to FichierAudio model

Closes IA-437

See merge request !24

  - Add parent_uuid and media_uuid fields to FichierAudio and ImageDocument models
  - Create MultimediaAPIService with dynamic UUID discovery for audio and images
  - Update serve_media_file to proxy both audio and image files through multimedia download API
  - Modify upload services to automatically store multimedia UUIDs during file upload
  - Centralize multimedia API URL configuration in constants.py

  - Create service to discover multimedia folder structure
  - Update audio serving to use multimedia download endpoint

- Add file_uuid field to store S3 file references

Closes IA-437

See merge request !23

- Updated French comments and error messages to English for better accessibility.
- Removed unnecessary logging statements and cleaned up code formatting.
- Enhanced error handling and response messages for audio file uploads and transcription processes.
- Improved logic for transcription processing, ensuring clearer flow and handling of options.
- Added detailed comments to clarify the purpose of various code sections.

  - Ajouter l'import correct d'ImageDocumentService depuis le module services
  - Supprimer la classe ImageDocumentService locale qui causait des conflits
  - Corriger l'AttributeError où la méthode upload_image_to_s3 n'était pas trouvée

  - Dynamic folder lookup (media->audio->user_id)
  - Fix API endpoints and status code handling (201)

Add unique filename generation for audio uploads and implement media file serving endpoint

Closes IA-437

See merge request !22

Add 'path' and 'interaction_id' fields to OCR batch API response

Closes IA-437

See merge request !21

  - Add 'path' and 'interaction_id' fields to transcription API response
  - Fix WindowsPath JSON serialization error by converting paths to strings
  - Update image_document path generation to avoid database duplicates

- Refactor FichierAudioService to store user uploads in media/audio/uploads/{user_id}/

Closes IA-437

See merge request !20

  - Add discussion_id field to FichierAudio model with index
  - Create migration 0004 for both model changes
  - Support optional discussion_id for linking media to conversations
  - Update image upload API to accept and store discussion_id
  - Update speech-to-text API to accept and store discussion_id
  - Update OCR batch processing to handle discussion_id
  - Add discussion_id to API responses for traceability

  - Refactor TextToSpeechService to store generated audio in media/audio/generated/temp/
  - Add permanent TTS storage in media/audio/generated/permanent/{user_id}/
  - Add new endpoint POST /api/synthesis/{id}/save for permanent TTS storage
  - Update SpeechToTextService for backward compatibility with old audio paths
  - Create separate folders to avoid confusion between user uploads and TTS files

Resolve IA-437 "Feature/"

Closes IA-437

See merge request !19

Update CHANGELOG for version 5.3.0 and change BRAINBOOST_GESTION_MODELES_URL to use secure HTTPS endpoint

Resolve IA-437 "Feature/"

Closes IA-437

See merge request !18

 add brainboost-traitement-multimedias-service and brainboost-api-rec.mc-intern.com in allowed hosts

Resolve IA-437 "Feature/"

Closes IA-437

See merge request !16

  Updated OCR service to resolve list index out of range error and API parameter incompatibility issues. Changed OCR models endpoint to use correct API path and updated OpenAI API call